A Universal Deep Neural Network for In-Depth Cleaning of Single-Cell RNA-Seq Data

AbstractSingle cell RNA sequencing (scRNA-Seq) has been widely used in biomedical research and generated enormous volume and diversity of data. The raw data contain multiple types of noise and technical artifacts and need thorough cleaning. The existing denoising and imputation methods largely focus on a single type of noise (i.e. dropouts) and have strong distribution assumptions which greatly limit their performance and application. We designed and developed the AutoClass model, integrating two deep neural network components, an autoencoder and a classifier, as to maximize both noise removal and signal retention. AutoClass is free of distribution assumptions, hence can effectively clean a wide range of noises and artifacts. AutoClass outperforms the state-of-art methods in multiple types of scRNA-Seq data analyses, including data recovery, differential expression analysis, clustering analysis and batch effect removal. Importantly, AutoClass is robust on key hyperparameter settings including bottleneck layer size, pre-clustering number and classifier weight. We have made AutoClass open source at: https://github.com/datapplab/AutoClass.

Download Full-text

SINC: a scale-invariant deep-neural-network classifier for bulk and single-cell RNA-seq data

Bioinformatics ◽

10.1093/bioinformatics/btz801 ◽

2019 ◽

Vol 36 (6) ◽

pp. 1779-1784 ◽

Cited By ~ 1

Author(s):

Chuanqi Wang ◽

Jun Li

Keyword(s):

Neural Network ◽

Single Cell ◽

Count Data ◽

Deep Neural Network ◽

Sequencing Depth ◽

Supplementary Information ◽

Neural Network Classifier ◽

Rna Seq ◽

Scale Invariant ◽

Downstream Analysis

Abstract Motivation Scaling by sequencing depth is usually the first step of analysis of bulk or single-cell RNA-seq data, but estimating sequencing depth accurately can be difficult, especially for single-cell data, risking the validity of downstream analysis. It is thus of interest to eliminate the use of sequencing depth and analyze the original count data directly. Results We call an analysis method ‘scale-invariant’ (SI) if it gives the same result under different estimates of sequencing depth and hence can use the original count data without scaling. For the problem of classifying samples into pre-specified classes, such as normal versus cancerous, we develop a deep-neural-network based SI classifier named scale-invariant deep neural-network classifier (SINC). On nine bulk and single-cell datasets, the classification accuracy of SINC is better than or competitive to the best of eight other classifiers. SINC is easier to use and more reliable on data where proper sequencing depth is hard to determine. Availability and implementation This source code of SINC is available at https://www.nd.edu/∼jli9/SINC.zip. Supplementary information Supplementary data are available at Bioinformatics online.

Download Full-text

DeepDRIM: a deep neural network to reconstruct cell-type-specific gene regulatory network using single-cell RNA-Seq Data

10.1101/2021.02.03.429484 ◽

2021 ◽

Author(s):

Jiaxing Chen ◽

Chinwang Cheong ◽

Liang Lan ◽

Xin Zhou ◽

Jiming Liu ◽

...

Keyword(s):

Neural Network ◽

Single Cell ◽

Regulatory Networks ◽

Deep Neural Network ◽

Neighborhood Context ◽

Cellular Heterogeneity ◽

Specific Gene ◽

Rna Seq ◽

Cell Type Specific ◽

Gene Regulatory

AbstractSingle-cell RNA sequencing is used to capture cell-specific gene expression, thus allowing reconstruction of gene regulatory networks. The existing algorithms struggle to deal with dropouts and cellular heterogeneity, and commonly require pseudotime-ordered cells. Here, we describe DeepDRIM a supervised deep neural network that represents gene pair joint expression as images and considers the neighborhood context to eliminate the transitive interactions. Deep-DRIM yields significantly better performance than the other nine algorithms used on the eight cell lines tested, and can be used to successfully discriminate key functional modules between patients with mild and severe symptoms of coronavirus disease 2019 (COVID-19).

Download Full-text

DeepImpute: an accurate, fast, and scalable deep neural network method to impute single-cell RNA-seq data

Genome Biology ◽

10.1186/s13059-019-1837-6 ◽

2019 ◽

Vol 20 (1) ◽

Cited By ~ 31

Author(s):

Cédric Arisdakessian ◽

Olivier Poirion ◽

Breck Yunits ◽

Xun Zhu ◽

Lana X. Garmire

Keyword(s):

Neural Network ◽

Single Cell ◽

Deep Neural Network ◽

Mean Squared Error ◽

Single Cells ◽

Rna Seq ◽

Squared Error ◽

Study Gene Expression ◽

Network Method ◽

The Mean

Abstract Single-cell RNA sequencing (scRNA-seq) offers new opportunities to study gene expression of tens of thousands of single cells simultaneously. We present DeepImpute, a deep neural network-based imputation algorithm that uses dropout layers and loss functions to learn patterns in the data, allowing for accurate imputation. Overall, DeepImpute yields better accuracy than other six publicly available scRNA-seq imputation methods on experimental data, as measured by the mean squared error or Pearson’s correlation coefficient. DeepImpute is an accurate, fast, and scalable imputation tool that is suited to handle the ever-increasing volume of scRNA-seq data, and is freely available at https://github.com/lanagarmire/DeepImpute.

Download Full-text

Imputing Single-cell RNA-seq data by combining Graph Convolution and Autoencoder Neural Networks

10.1101/2020.02.05.935296 ◽

2020 ◽

Cited By ~ 3

Author(s):

Jiahua Rao ◽

Xiang Zhou ◽

Yutong Lu ◽

Huiying Zhao ◽

Yuedong Yang

Keyword(s):

Single Cell ◽

Clustering Analysis ◽

State Of The Art ◽

Differential Expression Analysis ◽

Gene Interactions ◽

Rna Seq ◽

Sequencing Technology ◽

Imputation Methods ◽

Downstream Analysis ◽

Low Dimensional

AbstractSingle-cell RNA sequencing technology promotes the profiling of single-cell transcriptomes at an unprecedented throughput and resolution. However, in scRNA-seq studies, only a low amount of sequenced mRNA in each cell leads to missing detection for a portion of mRNA molecules, i.e. the dropout problem. The dropout event hinders various downstream analysis, such as clustering analysis, differential expression analysis, and inference of gene-to-gene relationships. Therefore, it is necessary to develop robust and effective imputation methods for the increasing scRNA-seq data. In this study, we have developed an imputation method (GraphSCI) to impute the dropout events in scRNA-seq data based on the graph convolution networks. The method takes advantage of low-dimensional representations of similar cells and gene-gene interactions to impute the dropouts. Extensive experiments demonstrated that GraphSCI outperforms other state-of-the-art methods for imputation on both simulated and real scRNA-seq data. Meanwhile, GraphSCI is able to accurately infer gene-to-gene relationships by utilizing the imputed matrix that are concealed by dropout events in raw data.

Download Full-text

Clustering Single-Cell RNA-Seq Data with Regularized Gaussian Graphical Model

Genes ◽

10.3390/genes12020311 ◽

2021 ◽

Vol 12 (2) ◽

pp. 311

Author(s):

Zhenqiu Liu

Keyword(s):

Single Cell ◽

Free Parameter ◽

Graphical Model ◽

Expression Patterns ◽

Information Criterion ◽

Log P ◽

Rna Seq ◽

Clustering Methods ◽

Wide Range ◽

Free Parameters

Single-cell RNA-seq (scRNA-seq) is a powerful tool to measure the expression patterns of individual cells and discover heterogeneity and functional diversity among cell populations. Due to variability, it is challenging to analyze such data efficiently. Many clustering methods have been developed using at least one free parameter. Different choices for free parameters may lead to substantially different visualizations and clusters. Tuning free parameters is also time consuming. Thus there is need for a simple, robust, and efficient clustering method. In this paper, we propose a new regularized Gaussian graphical clustering (RGGC) method for scRNA-seq data. RGGC is based on high-order (partial) correlations and subspace learning, and is robust over a wide-range of a regularized parameter λ. Therefore, we can simply set λ=2 or λ=log(p) for AIC (Akaike information criterion) or BIC (Bayesian information criterion) without cross-validation. Cell subpopulations are discovered by the Louvain community detection algorithm that determines the number of clusters automatically. There is no free parameter to be tuned with RGGC. When evaluated with simulated and benchmark scRNA-seq data sets against widely used methods, RGGC is computationally efficient and one of the top performers. It can detect inter-sample cell heterogeneity, when applied to glioblastoma scRNA-seq data.

Download Full-text

Single-cell conventional pap smear image classification using pre-trained deep neural network architectures

BMC Biomedical Engineering ◽

10.1186/s42490-021-00056-6 ◽

2021 ◽

Vol 3 (1) ◽

Author(s):

Mohammed Aliy Mohammed ◽

Fetulhak Abdurahman ◽

Yodit Abebe Ayalew

Keyword(s):

Neural Network ◽

Cervical Cancer ◽

Computer Vision ◽

Single Cell ◽

Deep Neural Network ◽

Deep Neural Networks ◽

Pap Smear ◽

Experimental Result ◽

Network Architectures ◽

Average Accuracy

Abstract Background Automating cytology-based cervical cancer screening could alleviate the shortage of skilled pathologists in developing countries. Up until now, computer vision experts have attempted numerous semi and fully automated approaches to address the need. Yet, these days, leveraging the astonishing accuracy and reproducibility of deep neural networks has become common among computer vision experts. In this regard, the purpose of this study is to classify single-cell Pap smear (cytology) images using pre-trained deep convolutional neural network (DCNN) image classifiers. We have fine-tuned the top ten pre-trained DCNN image classifiers and evaluated them using five class single-cell Pap smear images from SIPaKMeD dataset. The pre-trained DCNN image classifiers were selected from Keras Applications based on their top 1% accuracy. Results Our experimental result demonstrated that from the selected top-ten pre-trained DCNN image classifiers DenseNet169 outperformed with an average accuracy, precision, recall, and F1-score of 0.990, 0.974, 0.974, and 0.974, respectively. Moreover, it dashed the benchmark accuracy proposed by the creators of the dataset with 3.70%. Conclusions Even though the size of DenseNet169 is small compared to the experimented pre-trained DCNN image classifiers, yet, it is not suitable for mobile or edge devices. Further experimentation with mobile or small-size DCNN image classifiers is required to extend the applicability of the models in real-world demands. In addition, since all experiments used the SIPaKMeD dataset, additional experiments will be needed using new datasets to enhance the generalizability of the models.

Download Full-text

Leveraging high-powered RNA-Seq datasets to improve inference of regulatory activity in single-cell RNA-Seq data

10.1101/553040 ◽

2019 ◽

Cited By ~ 1

Author(s):

Ning Wang ◽

Andrew E. Teschendorff

Keyword(s):

Transcription Factors ◽

Single Cell ◽

Cell Fate ◽

Regulatory Networks ◽

Large Scale ◽

Single Cells ◽

Differential Expression Analysis ◽

Dropout Rate ◽

Rna Seq ◽

Regulatory Activity

AbstractInferring the activity of transcription factors in single cells is a key task to improve our understanding of development and complex genetic diseases. This task is, however, challenging due to the relatively large dropout rate and noisy nature of single-cell RNA-Seq data. Here we present a novel statistical inference framework called SCIRA (Single Cell Inference of Regulatory Activity), which leverages the power of large-scale bulk RNA-Seq datasets to infer high-quality tissue-specific regulatory networks, from which regulatory activity estimates in single cells can be subsequently obtained. We show that SCIRA can correctly infer regulatory activity of transcription factors affected by high technical dropouts. In particular, SCIRA can improve sensitivity by as much as 70% compared to differential expression analysis and current state-of-the-art methods. Importantly, SCIRA can reveal novel regulators of cell-fate in tissue-development, even for cell-types that only make up 5% of the tissue, and can identify key novel tumor suppressor genes in cancer at single cell resolution. In summary, SCIRA will be an invaluable tool for single-cell studies aiming to accurately map activity patterns of key transcription factors during development, and how these are altered in disease.

Download Full-text